Author Topic: LiFePO4 Fuel Gauge for higher capacity 20Ahr-40Ahr, multi-cell batteries ??  (Read 2107 times)

0 Members and 1 Guest are viewing this topic.

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
Hi
I need a fuel gauge for  4S (4 cell series ~ 12V) LiFePO4 batterys with capacities in the range 20Ahr to 40Ahr.  The batteries are off-the-shelf traction type designed for deep discharge.

I have been looking at fuel gauge ICs.  Most are for single cell applications.  Some don't do LiFePO4 chemistry.  I can't find one that will do higher capacity batteries.

Any suggestions ??

Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline MadScientist

  • Frequent Contributor
  • **
  • Posts: 439
  • Country: 00
The best way is a micro controller programmed up , an attiny would be ideal. You really need to factor in Pukerts law , even if it’s close to 1 for Li chemistry .  You need to decide if you want just battery level monitoring or cell level monitoring as well
EE's: We use silicon to make things  smaller!
 

Offline aldi

  • Contributor
  • Posts: 28
  • Country: nl
I've some experience with the BQ40Z60 from TI (http://www.ti.com/product/BQ40Z60). It supports up to 4S / 25V, so it think it fits the bill. It takes some time to understand and tweak all the parameters, but when you've tweaked all the parameters to perfection it can be a nice, accurate IC.
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
I've some experience with the BQ40Z60 from TI (http://www.ti.com/product/BQ40Z60). It supports up to 4S / 25V, so it think it fits the bill. It takes some time to understand and tweak all the parameters, but when you've tweaked all the parameters to perfection it can be a nice, accurate IC.

Thanks.
It took a lot of digging but this device will work to 32A.Hr battery capacity. Good enough for my application.
It also includes a lot more features that I will ever use.
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline ttyridal

  • Contributor
  • Posts: 16
  • Country: no
You also have BQ34110 and BQ34Z100-G1 which looks very nice and quite simple to use.
 

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1265
  • Country: nz
I’m part way through a BQ50Z50-R2 design, and one thing to be aware of is that it takes quite a lot of programming (I.e. there are many settings).

One of these is the Chemistry ID, which you find by doing discharge testing of a sample cell and submitting a test log to the TI website. This can take some time to do and there’s a chance that TI won’t have a Chemistry ID, so I suggest you get onto it pretty quickly. (I think if they don’t have a Chemistry ID they could possibly make one, but i suspect that involves weeks of testing.)
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
I’m part way through a BQ50Z50-R2 design, and one thing to be aware of is that it takes quite a lot of programming (I.e. there are many settings).

One of these is the Chemistry ID, which you find by doing discharge testing of a sample cell and submitting a test log to the TI website. This can take some time to do and there’s a chance that TI won’t have a Chemistry ID, so I suggest you get onto it pretty quickly. (I think if they don’t have a Chemistry ID they could possibly make one, but i suspect that involves weeks of testing.)
Hi
That's a problem.  I am using off-the-shelf 4S batteries.  I don't have access to single cells. 
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
You also have BQ34110 and BQ34Z100-G1 which looks very nice and quite simple to use.
These look like good candidates.
These are very similar devices. It almost looks like two different design teams have been given the same requirements.  A shallow skim through the datasheets shows not much difference.  They are about the same age and price.   The BQ34110PW is available from more suppliers.

I am going to have to read the data sheets in detail to figure out where the differences are.  I expect the devil is in the detail.   
« Last Edit: March 25, 2019, 05:15:12 am by dazz1 »
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
I have reached the conclusion that the BQ34110 is an updated version.   
It has some improvements.  It is geared to rarely discharged battery packs (eg UPS) but that doesn't make it unsuitable for other applications.

It is the one I will use.
« Last Edit: March 25, 2019, 08:26:26 am by dazz1 »
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
Hi
Hit a snag.
The official TI adapter to program the Fuel Gauge is here: https://nz.mouser.com/ProductDetail/Texas-Instruments/EV2400/?qs=bdKHSS5iKg4yCqIkH9l8SQ%3D%3D&gclid=CjwKCAjw-OHkBRBkEiwAoOZqlxsU27RfUZL4bM2zgK9gYr_JqhwBS2O4AHVbcn99u7CwTiVOD5fQnxoCEL8QAvD_BwE

ONLY  $USD338 !!!!  :(

It's only a USB to I2C interface so that should be easy to find.  It also includes software to program the chips. 
I found a link to the "free" BQ software :  http://www.ti.com/tool/bqStudio  but the link doesn't work.

I found BQ studio here:  https://github.com/mhassaan/bqStudio



The actual evaluation module https://www.digikey.com/products/en?mpart=BQ34110EVM-796&v=296
is a much more reasonable  $USD103, but it needs the USB/I2C interface as well.
« Last Edit: March 25, 2019, 08:38:39 am by dazz1 »
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 

Offline jbb

  • Super Contributor
  • ***
  • Posts: 1265
  • Country: nz
Re series packs: you can test 4 series cells on 1 go, no problem.

Re the communications box: I had limited time so I just got the company to pay.  I think the schematics for the box are available so you might be able to make your own.
 

Offline dazz1Topic starter

  • Frequent Contributor
  • **
  • Posts: 851
  • Country: nz
Re the communications box: I had limited time so I just got the company to pay.  I think the schematics for the box are available so you might be able to make your own.

My problem is that I am the company and I don't really want to pay ~$USD500 to try out a $5 IC. 
Yes I could make the comms adapter but it needs firmware that TI don't publish.

They do publish the comms protocol so it wouldn't be too difficult to use alternative software/hardware to do the job.
Dazz

Over Engineering: Why make something simple when you can make it really complicated AND get it to work?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf